Labor policies and the configuration of the job market in Bolivia:

An analysis through the effects on the search and matching process

Ricardo Nogales1
Pamela Córdova2
Manuel Urquidi 3
Brisa Rejas4

Center for Research in Economics and Management

Universidad Privada Boliviana

August, 2015

Abstract

In this paper we present an analytical framework for assessing relations between

labor policies and the configuration of the labor market in Bolivia. We take into
account the existence of a large informal sector building on a job search and
matching partial-equilibrium model. In this framework, the informal sector is
defined as a part of the labor market that is not directly affected by labor policies
and not as an inferior residual of the formal sector. The model is calibrated and
solved numerically to reproduce the main characteristics of the labor market in
2013 and then it is used to simulate counterfactual scenarios. We focus on the
relations between three of the most recent labor policies in Bolivia, namely the
14th salary, minimum wage increases and contributions to ‘solidary’ pension funds,
and the sorting of workers between the formal and informal sectors, as well as
unemployment. We prove that the first and second policies increase wages in the
formal sector and that the third policy reduces them. We also prove that all three
policies reduce wage inequality within the formal sector.

Keywords: Labor Markets, Matching Models, Informality, Labor Policies.

1. Introduction

The labor market in Bolivia has been analyzed extensively from different
perspectives (Andersen & Muriel, 2007; Jiménez, 2003; Mercado et al., 2003;
Muriel & Jemio, 2010; Román, 2011). Understanding the mechanisms of the labor
market is interesting in its own right, but it is particularly important because these
mechanisms depict channels through which public policies and macroeconomic
conditions are passed on to people’s livelihoods. This may be one of the main

1 rnogales@upb.edu, Center for Research in Economics and Management,

Universidad Privada Boliviana
2 pcordova@upb.edu, Center for Reserach in Economics and Management,

Universidad Privada Boliviana

3 manuelu@aidb.org, XXX, Interamerican Development Bank
4 brejas@upb.edu, Center for Research in Economics and Management,

Universidad Privada Boliviana

1
reasons for labor policies in Bolivia to be very protective of workers’ rights.
Current policies in the country aim at promoting indefinite contracts, penalizing
dismissal, reducing redundancy, setting minimum wages and establishing
contributions to short-term and long-term social security, among other benefits for
workers.

The natural demographic process in Bolivia has yielded a stable increase of the
population of working age, but the dynamics of demand for workers in the formal
sector has not proven capable to follow this evolution (Román, 2011). Historically,
Bolivia is considered to have one of the largest informal sectors in Latin America
(Heckman & Pagés, 2000, 2004; ILO, 2013); by 2013, according to Household
Survey, the size of the informal sector may range from 74% to 91% according the
definition of an informal worker (affiliation with the pension system for the first
figure or the firm is deprived of tax identification number for the latter figure).
According to Román (2011), one reason for such an important size of the informal
sector may be the fact that while aiming to create better conditions for workers,
the labor policies that we mentioned earlier also imply costs for the employers. In
this sense, the size of the informal sector would be the result of attempts to avoid
at least some of these labor costs, which of course ends up depriving a
considerable part of the workforce of at least some of the benefits and legal rights
that come along with a formal labor contract.

Under these circumstances, analyzing the configuration of the labor market in

Bolivia and its relation to labor policies is not a straightforward issue; traditional
indicators such as unemployment and participation rates depict incomplete figures
that tend to blur the assessment of market efficiency and workers’ welfare, as well
as the effects of labor policies. We make the case that when analyzing the
configuration of the labor market in Bolivia, informality has to be taken into
account with particular attention; one may not rely on dual labor market theories
(see e.g. Saint-Paul, 1996), where the informal sector is conceived as a parallel,
illegal or excluded segment of the labor market. As stressed in Albrecht et al
(2008), Maloney (2004), Gong et al. (2004) and Mondragon & Peña (2008),
informal sectors of labor markets in Latin America may be interpreted as micro-
entrepreneurial and self-employment sectors where labor costs are sought to be
avoided and thus labor policies do not generate direct effects, rather than merely
inferior or disadvantaged residuals of the formal sectors, with no need for further
consideration.

In fact, according to Andersen & Muriel (2007), working in the informal sector may
actually be, to some extent, the reflection of individual or household preferences in
Bolivia. These preferences may be motivated by uncertainty about the future,
strong perception of gains derived for current wealth or lack of trust in the pension
system (Heckman & Pagés, 2004). A comparison of data coming from 2001 and
2013 Household Surveys may provide further support for this statement: the size
of the informal sector has remained practically invariant in this span of time
(Figure 1), whereas the part of workers receiving payments below the (real)
minimum wage has decreased and people receiving work payments between one
and two (real) minimum wages has increased (Figure 2. Perhaps this may indicate

2
that the informal sector has been absorbing workers with intermediate levels of
productivity.

We defend the idea that the informal sector is an important source of income for
many households in the country and thus it has a crucial social relevance.
However, there is also vast empirical evidence about the possibility of a negative
relation between aggregate productivity or aggregate output and the size of the
informal sector (see e.g. Leal-Ordóñez, 2014), as well as aggregate education levels
and the size of the informal sector (see e.g. Mastalioglu & Rigolini, 2006). These
stylized facts, combined with deprivation of at least some labor benefits for
workers, support the general perception of informal jobs as ‘low quality positions’
(World Bank, 2009).

Figure 1: Evolution of informality in period 2000-2013 defined as the part of the

workers with no affiliation to the pension system

100 84.9 82.2

80.2
72.4 72.5 68.2 72.2 74.1 73.5 74.5
80

60

40 27.8
25.5
20.1 22.2 22.4 19.2 21.8 26.5
15.6 14.6
20

0
2000 2001 2002 2003 2006 2007 2008 2009 2011 2012

Source: Household Surveys, National Institute of Statistics

FALTARIA COMPLETAR 2013

QUIZA SERIA INTERSANTE TENER SOLO LA EVOLUCION CONJUNTA.

3
Figure 2: Distribution of real wages in period 1999-2013
100%
13% 12%
90% 23% 23% 18%
30% 29% 30% 27% 26% 6% 5%
80% 44% 8% 13%
11% 9% 16%
70% 11% 11%
10% 11% 12% 20%
60%
18% 21%
19% 17% 17% 18%
50% 13% 19%
46% 51%
40%
16% 19% 19% 27% 39%
30% 19% 26% 30%
31%
20% 9%

10% 23% 24% 20% 18% 21% 20% 18%

18% 13% 16% 15%
0%
1999 2001 2002 2005 2006 2007 2008 2009 2011 2012 2013

Menor a WMin Entre 1 y 2 WMin Entre 2 y 3 WMin

Entre 3 y 4 WMin Más que 5 WMin

Source: Household Surveys, National Institute of Statistics

FALTARIA PONER LA LEYENDA EN INGLES

We argue that this complex context calls for a thorough analysis of the relation
between labor policies and the job market configuration in the country, in the
quest for effective policymaking towards improvement of employment conditions.
In this paper, we aim to make a contribution to this matter, investigating the
effects of these policies on the sorting of workers between unemployment and the
formal and informal sectors. Furthermore, we assess their effect on formal wages
and on wage inequality within the formal sector. We focus on three of the most
recent labor policies in Bolivia, namely the conditional payment of a 14th salary,
the increase of the minimum wage between in 2013 and the compulsory
contributions to a pension fund called “Fondo Solidario”, in benefit of the least
favored workers in terms of their chances for adequate retirement conditions.

Our strategy consists of developing an adapted version of a job search and

matching model in an economy with a large informal sector proposed by Albrecht
et al. (2008), which in turn builds upon job market frictions analyses by Diamond
(1982) and Mortensen & Pissarides (1994). According to the discussion above, we
stress that the informal sector is an important part of the labor market where
contracts and self-employment decisions are not framed within rules established
by public policies that are designed to directly affect formal labor contracts. We
pay particular attention to arrive at an appropriate calibration of this analytical
framework in order for it to be useful for policy purposes. For this, we draw
inspiration from similar versions of this partial equilibrium model have been used
for assessing the role of public policies on the configuration of labor markets in

4
Colombia (Robayo, 2014), Brazil (Bosch, 2006) and Mexico (Bosch & Esteban-
Pretel, 2006). Once it is appropriately calibrated, this model allows us gauge the
effects of public policies on the configuration of the labor market through
simulation of counterfactual scenarios.

The document is structured as follows: Section 2 presents a brief overview of the

labor market in Bolivia, including its configuration and current labor policies.
Section 3 presents the essential characteristics and mechanisms of our job search
and matching model, as well as the calibration of its parameters. Section 4 is the
crux of document, as it presents counterfactual analyses to discuss the effects of
the three labor policies that are the focus of this study. Finally, Section 5 presents
our main concluding remarks.

2. A brief overview of the labor market in Bolivia

As in many Latin American countries, in Bolivia coexist both low levels of

unemployment and high levels of informality. As shown in Figure 3, the mean
annual unemployment rate in the country in period 2000-2013 is 6.3%, which is
lower than the average of South America in the same span of time (9.3%). This rate
shows a persistently decreasing trend that has accelerated in recent years; in
capital cities, the average annual reduction has gone from -0.25 percentage points
in period 1989-1999 to -0.48 points in period 2000-2012. The dynamics of
unemployment rates in the country are driven by a significant reduction of the
severance rate (-0.43 percentage points per year in the latter period; see Figure 4).
At the same time, participation rates and employment-to-population ratios have
been increasing by effect of natural demographic dynamics, which means that even
if the potential workforce is steadily increasing, it has become easier for workers
to find another source of income should they come to lose their position.

Figure 3. Average unemployment rates in South America (2000-2013)

Colombia 13.3
Argentina 11.7
Uruguay 10.2
Venezuela 10.0
South America 9.3
Brazil 9.2
Chile 9.0
Paraguay 7.8
Ecuador 7.6
Bolivia (Plurinational State of) 6.3
Perou 5.6
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0

Source: Economic Commission for Latin America and the Caribbean – ECLAC

Figure 4: Evolution of unemployment rates, severance rates and first-time

applicants unemployment rates in Bolivia

5
 10.0

8.0

6.0

4.0

2.0

0.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2008 2011 2012
(p) (p)

Unemployment rate
Severance rate
First time applicants unemployment rate

Source: Unidad de Análisis de Políticas Económicas y Sociales - UDAPE

It is clear, however, that without a thorough assessment of the conditions of

employment, low unemployment rates alone do not allow to deduce that the labor
market has gained efficiency or that public policies aiming at improving workers’
livelihoods have been effective. According to the 2013 official Household Survey,
most of the Bolivian workforce is employed under informal conditions in the sense
that they are not affiliated with the pension funds (73%). Taking this definition for
informality, on average, this sector has represented 81% of the labor market
between 1999 and 2012.

The lack of long-term social protection is usually combined with deprivation of

other social security rights, such as dismissal compensations or access to health
insurance, which has often led to perceive informal jobs as less advantaged
working positions compared to formal ones. Indeed, as we have mentioned earlier,
it has been empirically proven that informality is associated with low productivity
thus it may considered as a working condition for people with fewer to be
competitive in the formal sector due to their levels of skills and qualifications
(Taymaz, 2009; World Bank, 2009). By 2013, the most part of college
postgraduates were formal salaried workers (79%, according to the Household
Survey).

The causes of such a high levels of informality in Bolivia are complex and their
precise identification is beyond the scope of this study; however, in Román (2011)
it is argued that the possibility to avoid costs associated with the creation of formal
position may be one of the main drivers of such a clear predominance of the
informal sector in the economic activity. It is worth mentioning that, according to
the Household Survey, 60% of the workers are self-employed. Furthermore, the
rapid increase of the creation rate of ‘tiny’ business (i.e. owned by only one
person) in capital cities, which is depicted in Figure 5, may also provide support
for this statement. Legal requirements for the establishment of these tiny-scale
firms are the least restrictive in terms of fees and time and they are the only type

6
of firms recognized by law that may not have registration as formal employers
before the Ministry of Labor5.

Figure 5: Evolution of number of firms in capital cities

70,000

60,000

50,000
Numer of firms

40,000

30,000

20,000

10,000

0
2005 2006 2007 2008 2009 2010 2011 2012 2013

Firms owned by one person Limited liability companies

Coporations Others

Source: Unidad de Análisis de Políticas Económicas y Sociales - UDAPE

Indeed, the gains of setting up this type of business instead of starting a larger
business that requires registration as formal employer may appear quite
attractive. According to the World Bank (2015), starting a larger-scaled business in
the country implies undertaking 15 procedures that take 49 days long and
represent costs equivalent to 65% of the average income per capita; these figures
are well above the average of the region. Thus between 2012 and 2013, more than
38’000 new tiny businesses have been listed in official public records6; by the end
of 2013, these tiny firms represented 94% of the officially established enterprises
operating in capital cities.

At this point, it is important to have a clear idea about the magnitude of the labor
costs that some agents in the labor market seek to avoid through informality.
These costs are established by set of labor market regulations that has been rather
stable over time, as it is are grounded on the General Law of Employment of 1942.
Later, the Social Security Code of 1956 and the current Pension Law established in
2010 have complemented the law of 1942. Muriel (2014) presents a thorough
review of labor legislation in Bolivia; in order to arrive at an adequate calibration
of our analytical framework, here we limit ourselves to present the essential
aspects of these laws that are related to costs for formal employers. The charges

5 See set of norms for registration of firms:

http://www.fundempresa.org.bo/registro-de-comercio-de-bolivia/marco-
normativo/
and Supreme Decree 288 of Sept. 2009.
6 The records of formally established commercial enterprises are managed by

Fundempresa: http://fundempresa.org.bo

7
that this legislation imposes to the employers include: compulsory contributions to
the national health system on behalf of the worker (10% of the salary) and to a
national program for subsidized housing (2% of the salary); the payment of an
insurance prime against personal risks at the working post (1.71% of the salary); a
compensation upon removal from post equivalent to one monthly salary for each
year at work7; a compensation in case of unfair dismissal equivalent to three
monthly salaries, and finally, a 13th salary every year. The formal benefits for
which we make a case thus far depict an increase of 30.37% of the yearly salary.

Since 2006, these benefits for workers have increased on the grounds of a
sustained macroeconomic performance and the search for a more equitable
redistribution of wealth between workers and employers, as well a reduction of
poverty (World Bank XXX FUENTE?). In this study we focus on three of these
benefits that have created new labor costs for the employers. The first benefit is
related to compulsory contributions to a pension fund called “Fondo Solidario”,
that seeks to help of the least favored workers in terms of their chances for
adequate retirement conditions; it is included in the Pension Law of 2010 and the
employers’ charge amounts to 10% of the salary. The second benefit is a gradual
increase of the minimum wage; in period 2001-2013, the minimum wage has
registered an average nominal increase of 10% per year, which corresponds to an
average real increase of 4.15% per year. Between 2012 and 2013 the minimum
wage has registered a nominal increase of 20% and a real increase of 13.5%. These
minimum wage increases affect all formal labor contracts through a scale of
compulsory seniority bonus that is indexed to the minimum wage8. Finally, the
third benefit is the payment of a 14th salary conditional to ‘good’ yearly
macroeconomic performance. Since 2013, Supreme Decree 1802 establishes that
this benefit becomes effective if aggregate economic growth exceeds 4.5%, a
condition that has been fulfilled in 2013 and 2014.

The facts that we briefly present here lead to perceive that there is a relation
between labor policies and the configuration of the labor market, and that taking
into account the informal sector is of paramount importance to identify and
understand this relation. We go on to investigate this issue building on a
framework where it possible to analyze decisions taken, on the one hand, by the
workers in terms of taking up an informal job as a source of income or compete for
a formal position and, on the other hand, by formal employers in terms of keeping
a worker in place or creating vacancies and choosing workers to fill them up. A job
search and matching model similar to the one proposed by Albrecht et al. (2008),
Diamond (1982) and Mortensen & Pissarides (1994) serves well for this purpose
and it allows to gauge the extent to which some of the labor policies that we
describe here affect individual decisions.

3. The Job Search and Matching Model

7 This compensation can even be made effective (normally every five years) while
the worker is still in place, through an agreement with the employer.
8 These bonus exist since 1985 and are established by Supreme Decree 21060.

8
We describe here the essential technical aspects of the partial-equilibrium model
that we use for assessing job market dynamics in Bolivia. We recall that in this
framework, the informal sector is conceived as an unregulated sector in the sense
that it is not directly affected by labor policies.

3.1. The Basic Setting

Based on the discussion above, we associate the informal sector with low levels of
productivity and thus, low levels of income flows for workers. In this sector, job
opportunities and job destructions arrive at exogenous Poisson rates denoted as 𝛼
and 𝛿, respectively. A job in this sector generates a constant exogenous flow of
income denoted as 𝑦0 . The reason for treating this sector with such simplicity is
the fact that the dynamics of creation and destruction of informal jobs cannot be
represented through a competitive process of job-matching governed by
optimizing behavior of firms and potential workers as it is proposed in
mainstream labor economics theory. We argue that this is particularly true in the
Bolivian case due to the predominance of self-employment in the informal sector,
thus creation and destruction rates may no be endogenous.

The dynamics in the formal sector of the job market are more complex. Details of
the model are presented in Albrecht et al. (2008); here, we will limit ourselves to
present and discuss the most relevant aspects of its formal conception and explain
how this model is adapted for the Bolivian case. Let us note 𝑦̅ the level of potential
productivity of a certain worker; when a firm encounters a potential worker, her
actual productivity cannot be known with certainty, thus we assume that the
values of 𝑦 are iid draws from a standard uniform distribution9, i.e. 𝑝𝑟𝑜𝑏[𝑦 = 𝑦̅] ≡
𝑓(𝑦̅) = 1, ∀𝑦̅ ∈ (0,1) and 𝑝𝑟𝑜𝑏[𝑦 ≤ 𝑦̅] ≡ 𝐹(𝑦̅) = 𝑦̅, ∀𝑦̅ ∈ (0,1)

In the formal sector, the rate of arrival of job opportunities is given by the
matching function. We draw inspiration from Pentrongolo & Pissarides (2011) to
propose this Cobb-Douglas form for the matching function: 𝑀(𝑈, 𝑅) = 𝐴𝑅𝛽 𝑈1−𝛽
where 𝑅 denotes the number of vacancies in this sector, 𝑈 denotes the stock of
unemployed workers, 𝛽 denotes the sensibility of the number of job matches to the
number of vacancies and 𝐴 denotes the level of technology in the matching
function. This latter parameter is associated to the efficiency of the matching
process in the labor market: greater values of 𝐴 depict better opportunities for an
unemployed worker to find a formal position. The number of matches available for
each unemployed worker define the rate at which job opportunities arrive:
𝑀(𝑈,𝑅) 𝑅
≡ 𝑚(𝜃) = 𝐴𝜃𝛽 , where 𝜃 ≡ 𝑈. This ratio depicts the market tightness; from
𝑈
the workers’ perspective, lower values of 𝜃 depict greater difficulties to find a
formal job. From the employer’s perspective, lower values of 𝜃 depict greater
choices to fill up a vacancy.

The job destruction process in the formal sector is endogenous as well; the model
considers the arrival of negative productivity shocks at an exogenous Poisson rate

9Albrecht et al. (2008) cover in detail the case in which levels of productivity are
governed by any other continuous distribution, with no alteration to qualitative
results that we present here.

9
𝜆. When a worker is occupied in a formal position, these shocks reduce the
productivity at which the worker actually performs and thus threaten the stability
of the match. From the firms’ perspective, there is a minimum level of productivity
that is required in order to be attractive to maintain a worker in place, which is
called the reservation productivity, denoted as 𝑅(𝑦̅). If the shock diminishes the
worker’s productivity to a level below 𝑅(𝑦̅), then the job is destroyed; if the shock
is not ‘strong enough’, the productivity is diminished to a new level 𝑦′ < 𝑦̅, but the
job is not destroyed. Therefore, 𝑦̅ may be understood as the workers’ ‘maximum’
or ‘potential’ level of productivity and 𝑦′ as her ‘actual’ or ‘observed’ level of
productivity.

The intensity of the productivity shocks cannot be known with certainty, but their
conditional probability of occurrence can be deduced from the distribution of
productivity levels. It is important to consider the fact that shocks can only
endanger the stability of the job, because from a technical perspective, this
assumption implies the values of the after-shock productivity are truncated. For a
certain level of productivity 𝑦̅ 𝜖 (0,1), the probability of a shock resulting in a
lower level of productivity 𝑦′ < 𝑦̅ is given by:

𝑝𝑟𝑜𝑏{𝑦 = 𝑦 ′ } 𝑓(𝑦 ′ ) 1
𝑝𝑟𝑜𝑏{𝑦 = 𝑦 ′ |𝑦′ ≤ 𝑦̅} = = = ∀ 0 ≤ 𝑦′ ≤ 𝑦̅ (1)
𝑝𝑟𝑜𝑏{𝑦 ≤ 𝑦̅} 𝐹(𝑦̅) 𝑦̅

Wages in the formal sector, denoted as 𝜔, are associated to each workers’ level of
potential productivity 𝑦̅ and depend on whether or not a shock has taken place
inducing a lower level of productivity 𝑦′, i.e. 𝜔 = 𝜔(𝑦′, 𝑦̅).

In this framework, we focus on two types of labor policies. The first are policies
that depict costs for the employer in case of dismissal; they represent severance
taxes and will be denoted as 𝑠. In the Bolivian case, the only policy of this type is a
compensation for unfair dismissal.

The second type of policies depicts costs for the employer while they are engaged
in a formal labor contract; they represent payroll taxes and will be denoted as 𝜏. In
turn, we propose to divide these policies according to their effect on salaries. We
reason that some of the current payroll taxes in Bolivia represent a direct increase
in salaries for formal workers, which is the case of the 13th and 14th salaries,
minimum wage increases and the compensation upon removal from post. In
contrasting fashion, some payroll taxes do not represent a direct benefit to the
worker in terms of current monetary compensations, such as compulsory
contributions to “Fondo Solidario” and to the national health system, as well as the
prime of insurance against personal risks at the working place or contributions to
the subsidized housing program. Taking into account the difference between these
two types of payroll taxes is particularly important because it determines whether
or not they bare a direct effect over payments to workers.

3.1.1. The search process: the workers’ perspective

For a worker of maximum productivity 𝑦̅, there are three possible states: i) to be
unemployed, which generates a future expected gain denoted as 𝑈(𝑦̅); ii) to be

10
employed in the informal sector, which generates a future expected gain denoted
as 𝑁0 (𝑦̅) and iii) to be employed in the formal sector, generating a future expected
gain denoted as 𝑁1 (𝑦 ′ , 𝑦̅)10.

We consider that if a worker is unemployed, she does not receive any kind of
monetary income because in Bolivia there are no public unemployment insurance
or conditional transfers in this situation. At this state, the current-valued expected
gain for the (potential) worker is given by:

𝑟𝑈(𝑦̅) = 𝛼 max[𝑁0 (𝑦̅) − 𝑈(𝑦̅), 0] + 𝑚(𝜃) max[𝑁1 (𝑦 ′ , 𝑦̅) − 𝑈(𝑦̅), 0] (2)

When a worker is employed in the formal sector, holding a position that offers all
social security rights, the current-valued expected gain for the worker is:

𝑟𝑁1 (𝑦 ′ , 𝑦̅) = 𝜔(𝑦 ′ , 𝑦̅) + 𝜆𝑝𝑟𝑜𝑏{𝑦 ≤ 𝑅(𝑦̅)|𝑦 ≤ 𝑦̅}(𝑈(𝑦̅) − 𝑁1 (𝑦 ′ , 𝑦̅))

𝑦̅

+ 𝜆 ∫ (𝑁1 (𝑥, 𝑦̅) − 𝑁1 (𝑦 ′ , 𝑦̅)) ∙ 𝑝𝑟𝑜𝑏{𝑦 = 𝑥|𝑦 ≤ 𝑦̅}𝑑𝑥

𝑅(𝑦̅)
𝑅(𝑦̅)
= 𝜔(𝑦 ′ , 𝑦̅) + 𝜆 (𝑈(𝑦̅) − 𝑁1 (𝑦 ′ , 𝑦̅))
𝑦̅
𝑦̅
1
+ 𝜆 ∫ (𝑁1 (𝑥, 𝑦̅) − 𝑁1 (𝑦 ′ , 𝑦̅)) 𝑑𝑥 (3)
𝑦̅
𝑅(𝑦̅)

When a worker is employed in the informal sector, she is deprived of some (or
perhaps all) social security rights. We consider that in this state workers can be
either self-employed or salaried. In this state, the current-valued expected gain for
the worker is:

𝑟𝑁0 (𝑦̅) = 𝑦0 + 𝛿(𝑈(𝑦̅) − 𝑁0 (𝑦̅)) (4)

3.1.2. The search process: the firms’ perspective

From the perspective of a formally established firm, there are two possible states.
First, a position is occupied by a worker of potential productivity 𝑦̅ and performs
at an actual productivity 𝑦 ′ . Let us denote as 𝐽(𝑦′, 𝑦̅) the expected gain for the firm
in this state. Second, the position is vacant giving an expected gain for the firm
denoted as 𝑉.

The current-valued expected gain in the first state is given by:

10Note that receiving the maximum wage is a particular case that occurs when
′
𝑦 = 𝑦.

11
 𝑟𝐽(𝑦 ′ , 𝑦̅) = 𝑦 ′ − 𝜔(𝑦 ′ , 𝑦̅)(1 + 𝜏) + 𝜆𝑝𝑟𝑜𝑏{𝑦 ≤ 𝑅(𝑦̅)|𝑦 ≤ 𝑦̅}(𝑉 − 𝐽(𝑦 ′ , 𝑦̅) − 𝑠)
𝑦̅

+ 𝜆 ∫ (𝐽(𝑥, 𝑦̅) − 𝐽(𝑦 ′ , 𝑦̅)) ∙ 𝑝𝑟𝑜𝑏{𝑦 = 𝑥|𝑦 ≤ 𝑦̅}𝑑𝑥

𝑅(𝑦̅)
𝑅(𝑦̅)
= 𝑦 ′ − 𝜔(𝑦 ′ , 𝑦̅)(1 − 𝜏) + 𝜆 (𝑉 − 𝐽(𝑦 ′ , 𝑦̅) − 𝑠)
𝑦̅
𝑦̅
1
+ 𝜆 ∫ (𝐽(𝑥, 𝑦̅) − 𝐽(𝑦 ′ , 𝑦̅)) ∙ 𝑑𝑥
𝑦̅
𝑅(𝑦̅)

In order to identify the current valued expected gain of a vacancy for a firm, let us
recall that the number of potential matches for each vacant post can be written as:
𝑀(𝑈,𝑉) 𝑚(𝜃)
≡ 𝜃 ; this ratio depicts the rate at which job opportunities arrive from the
𝑉
firm’s perspective. Thus the current-valued expected gain in this state is given by:

𝑚(𝜃)
𝑟𝑉 = −𝑐 + 𝐸{𝑚𝑎𝑥[𝐽(𝑦̅, 𝑦̅), 0]} (5)
𝜃

where 𝑐 represent the cost of keeping a vacancy.

After briefly describing the basic setting of the model, it is important to notice that
one key aspect of the framework proposed in Albrecht et al. (2008), which makes
it particularly suitable for assessing the Bolivian case, is that the maximum level of
productivity defines three types of workers that have very distinctive dynamics in
the labor market:

i) ‘Low productivity workers’ that are always bound to work in the

informal sector; their maximum level of productivity does not allow
them to aspire to occupy a formal position. Maloney (2004) and
attribute this to the fact that there are levels of qualifications that are
not sufficient for a worker to perceive that she is competitive enough in
the quest for a formal position and, in expectation, she would be better
off in the informal sector. Thus there is an productivity threshold, noted
𝑦 ∗ , from which workers having lower levels of productivity perceive
that the expected gain of being employed in the formal sector is
equivalent to the expected gain of being unemployed:

𝑁1 (𝑦, 𝑦) = 𝑈(𝑦) 𝑖𝑓 𝑦 ≤ 𝑦 ∗ (6)

ii) ‘High productivity workers’ that are never found in the informal sector.
If these workers were unemployed, their level of productivity allows
them to perceive that the opportunity-cost of leaving unemployment for
obtaining an informal source of income is too high. Thus there is a
second productivity threshold noted 𝑦 ∗∗ , from which for workers having
higher levels of productivity, the expected gain of occupying a formal
position is always higher than the expected gain of being in the informal
sector. This means that for a worker with a level of productivity 𝑦 ∗∗ , the

12
 expected gain of being unemployed is identical to the expected gain of
being in the informal sector:

𝑁0 (𝑦 ∗∗ ) = 𝑈(𝑦 ∗∗ ) < 𝑁1 (𝑦 ∗∗ , 𝑦 ∗∗ ) (7)

iii) ‘Medium productivity workers’ that may consider taking formal as well
as informal positions depending on which opportunity comes along first.
These workers have a level of productivity between 𝑦 ∗ and 𝑦 ∗∗ .

Thus in this framework, workers are divided in three groups according to the
values of the aforementioned thresholds 𝑦 ∗ and 𝑦 ∗∗ , which are endogenous and
may be influenced by labor policies.

3.2. The equilibrium and wage determination: a brief description

The labor market is in equilibrium if the market tightness, 𝜃, reaches a steady

value. Thus the fundamental steady state condition establishes that the expected
value of a vacancy is null, 𝑉 = 0, because when this condition is met, there is a
fixed number of vacancies offered by the firm (see Albrecht et al., 2008).

Of course, wages play a crucial role in labor market equilibrium and the fulfillment
of the fundamental steady-state condition. Wages in the formal sector are
determined through a Nash private bargaining process, in which the worker has an
exogenous negotiating power denoted as 𝛽, and that of the firm is denoted as 1 −
𝛽. The total gains of creating a formal job are calculated as the aggregation of the
gains for the worker and for the firm, each weighted by their respective bargaining
power.

For the identification of wage schedules compatible with equilibrium, two different
scenarios have to be considered: either i) the worker performs at maximum
productivity, i.e. no shocks have occurred, or ii) the worker performs at a
productivity level that is lower than her maximum potential. The main difference
between the two scenarios is that in the event of a productivity shock, the firm may
consider ending the job and thus has to decide whether to avoid or face severance
costs, 𝑠, established in Bolivian labor legislations. In formal terms, the gain for the
firm in the wage negotiation process depends on whether or not a productivity
shock has occurred. Thus wages schedules at equilibrium are determined through
a private negotiation process between the firm and the employer according to the
following program:

max[𝑁1 (𝑦 ′ , 𝑦̅) − 𝑈(𝑦̅)]𝛽 𝐽(𝑦 ′ , 𝑦̅)1−𝛽 𝑖𝑓 𝑦′ = 𝑦̅

𝜔(𝑦 ′ , 𝑦̅) 𝑠𝑜𝑙𝑣𝑒𝑠 { (8)
max[𝑁1 (𝑦 ′ , 𝑦̅) − 𝑈(𝑦̅)]𝛽 [𝐽(𝑦 ′ , 𝑦̅) + 𝑠]1−𝛽 𝑖𝑓 𝑦′ < 𝑦̅

At this point, a substantial modification to the model proposed by Albrecht et al.

(2008) is required. As we mentioned before, two of the labor policies that we
evaluate here, namely the 14th Salary and minimum wage increases, modify the
salary received by a formal worker but they are not subject to negotiation with the
employer. In a way, these policies represent a right to salary increases for formal
workers obtained through an exogenous collective bargaining process promoted

13
by public authorities. This bargaining process may not alter the optimization
program above, as it affects all formal labor contracts equally implying a transfer
of wealth from formal employers to formal workers. In presence of such policies,
the formal wage schedule at equilibrium is modified ex post factum to 𝜔 ̃(𝑦 ′ , 𝑦̅) as
following:

̃(𝑦 ′ , 𝑦̅) ≡ 𝜔(𝑦 ′ , 𝑦̅) ∙ (1 + 𝜉)

𝜔 (9)

where 𝜉 < 𝜏 is the part of the payroll taxes charged to the firm, 𝜏, that goes on to
increase the salary of formal workers.

3.3. Calibration

We calibrate the model to fit the main traits of the Bolivian urban labor market in
2013. We define this period in time as the baseline case of our model because we
rely on data coming from the 2013 Household Survey, which is the latest official
survey that is representative at the urban area. In order for the baseline case to
reproduce the main characteristics of the urban labor market as closely as
possible, it includes all current labor policies in the country, in particular the three
labor policies that are the focus of this study.

The time unit we choose to calibrate the model is a year; relying on Bolivian data
and previous studies (Albrecht et al., 2008; Robayo, 2014; Bosch, 2006; Shimer,
2005), we set the values for the parameters of the model as follows:

- The discount rate, 𝑟, is associated to the real interest rate, which is 4.8% for
2013 according to the World Bank.
- In the absence of longitudinal data suitable for the empirical estimation of a
matching function for the urban Bolivian labor market, we follow standard
results in the literature to set values for the parameters of this function:
o The elasticity of the number of matches in a year with respect to the
market tightness, noted 𝛽, is set to 0.5. Thus we assume that workers
and employers in the formal sector have equal negotiation power in
the wage determination process. In the absence of quantitative
information to propose alternative assumptions Albrecht et al.
(2008), Robayo (2014) and Shapiro & Sarzoza (2013), have also
resourced to the same value for this parameter.
o Robayo (2014) estimated that the level of the technology of the
matching function, 𝐴, lies within the range 1.5-2.2 in Colombia.
Bosch (2006) estimates similar values for 𝐴 for Mexico and Brazil.
We draw inspiration from the Colombian case to set 𝐴 to the lower
bound the estimated range, 1.5, because we intend to capture the
high levels of informality in Bolivia that we have discussed in
introductory paragraphs, in part, through a low value of this
parameter depicting a low level of efficiency of the matching process
in the formal sector.
- The income flow of informal workers, 𝑦0 , is set to 0.36, depicting the ratio
between the average income of workers that are not affiliated with the

14
 pension funds (a possible definition of informal workers) with low levels of
education and the 90th percent of urban formal wages11.
- We set the rate of arrival of productivity shocks, 𝜆, to 0.5. Let us mention
that according to the survival analysis theory (Cox & Oakes, 1984), the
separation rate in the formal sector can be expressed as the inverse of the
expected duration of formal employment. According to the Household
Survey, this duration is around 5 years in Bolivia, depicting a separation
rate of 0.2. We argue that the separation rate represents a lower bound for
the rate of arrival of productivity because these shocks may take place
without ending the match. Thus we draw inspiration from Albrecht et al.
(2008) to set the arrival rate to 0.5 arguing that this value has fit well for
assessing labor markets in Mexico, Colombia and Argentina.
- Similarly, the job destruction rate in the informal sector, 𝛿, can be
represented as the inverse of the duration of informal employment, which,
according to the Household Survey is around 8 years. Thus, we set 𝛿 = 0.12.
- The rate of arrival of employment opportunities in the informal sector, 𝛼,
has been set to 15 in order to replicate unemployment rates in the informal
sector. This value is high enough to se this rate near zero; we argue that this
value is reasonable for the Bolivian case because it is easy to find a source
of income in the informal sector, whether it may come from holding a
salaried position or being self-employment.
- The cost of a vacancy for the firm, 𝑐, is set to 0.4. We have set this value so
that the central steady state condition (𝑉 = 0) holds; we have aimed a
estimating as closely as possible the fact that when this condition is fulfilled,
𝑐 depicts the expected gain of a hiring a new worker though a formal
contract. We have associated this expected gain to the proportion of the
mean years of education in the total working population (9.6) with respect
to the maximum years of education registered in the Household Survey
(23).
- The severance tax, 𝑠, is set to 5%. We propose this value as an estimation of
the due compensation in case of unfair dismissal, taking into account that,
according to the Household Survey, the average duration of employment in
the formal sector is 5 years.
- The payroll taxes, 𝜏, may take into account the 13th salary (8.33%), the
prime for insurance against professional risks at work (1.71%), the
contribution the subsidized housing program (2%), the compensation for
removal from post (8.33%), the contribution to the national health system
(10%), the contribution to “Fondo Solidario” (3%), the 14th salary (8.33%)
and the minimum wage increase, which we set to 3.5%. This latter figure
has been calculated taking into account that, according to the Household
Survey, the average duration of employment in 5 years and that Supreme
Decree 21060 establishes a corresponding seniority bonus of 11% of the
salary.
- The taxes considered in 𝜏, with the exception of the prime for insurance
against professional risks at work, the contribution to the subsidized
housing program and the contribution to “Fondo Solidario”, have a direct

11This percentage has been calculated in order to avoid right-side outliers in the
formal wages distribution.

15
 impact on the workers’ salary. Thus all the other payroll taxes may be
considered in the coefficient that modifies the wage schedule to take into
account this impact, 𝜉.

4. Results and Simulations

In order to gauge the effects of the 14th salary, the contribution to “Fondo
Solidario” and minimum wage increases on the labor market configuration, we
compare the baseline case with counterfactual scenarios that simulate the absence
of each labor policy, one at a time. The results of these simulations are presented in
Table 1

The effect of minimum wage increases is assessed comparing the base scenario
with a counterfactual situation where there are no wage increases, while
everything else is held constant; we call this the counterfactual scenario 1. The
effect of the 14th salary is assessed comparing the latter scenario to another
counterfactual situation in which neither the wage increases not this additional
salary exist; we call this situation the counterfactual scenario 2. In a similar
manner, the effect of the contribution to “Fondo Solidario” is gauged comparing
the latter scenario with a situation in which neither of the three labor policies
exists; we call this situation counterfactual scenario 3. We propose to pursue this
strategy for impact assessment because it depicts a progressive reduction of
payroll taxes until a simulated scenario without labor policies.

Table 1: Model output for each scenario

Baseline Counter- Counter- Counter-
Scenario factual factual factual
(2013) Scenario 1 Scenario 2 Scenario 3
Labor policies 𝝉 45.20% 41.70% 33.37% 30.37%
configuration s 5.00% 5.00% 5.00% 5.00%
𝝃 28.49% 24.99% 16.66% 16.66%
Model Output 𝜽 1.24 1.25 1.23 1.24
(Selected) 𝒚∗ 0.5417 0.5292 0.4995 0.4888
𝒚∗∗ 0.8144 0.7942 0.7501 0.7328
𝑹(𝒚∗∗ ) 0.44 0.43 0.41 0.4
Workforce distribution Low 54.17% 52.92% 49.95% 48.88%
by type of worker Medium 27.27% 26.50% 25.06% 24.40%
High 18.56% 20.58% 24.99% 26.72%
Unemployment rates Low 0.79% 0.79% 0.79% 0.79%
by type of worker an Medium 0.76% 0.75% 0.76% 0.76%
total High 14.13% 14.11% 14.15% 14.13%
Total 3.26% 3.52% 4.12% 4.35%
Probability of informal Low 99.21% 99.21% 99.21% 99.21%
employment by type of Medium 94.53% 94.54% 94.48% 94.51%
worker High 0.00% 0.00% 0.00% 0.00%
Probability of formal Low 0.00% 0.00% 0.00% 0.00%
employment by type of Medium 4.72% 4.71% 4.76% 4.74%
worker High 85.87% 85.89% 85.85% 85.87%
Duration of unemployment (Months) 7.18 7.16 7.21 7.18

16
Matching Function Output 1.67 1.68 1.66 1.67
Formal wage Mean 0.582 0.5744 0.5548 0.5626
Stand. Dev. 0.0687 0.0708 0.0758 0.0799
Informal wage / Mean formal wage 0.640 0.619 0.627 0.649
Source: Own

In a general manner, one of the main results of our study is that all the considered
labor policies increase informality; we provide different kinds of evidence to back
up this affirmation. We make the case that the increase of informality is due to the
fact that the labor policies that we consider depict higher costs for formal
employers and thus result in a reduction of the rate at which vacancies in the
formal sector are created. We are able to distinguish the ‘type’ of informal workers
that have increased due to these policies: first, ‘workers with low productivity’
who abandon desires of competing for formal positions and second, ‘workers with
medium productivity’ who may consider taking up informal jobs should formal
jobs were not available to them. The first may be referred to as permanent
informal workers and the latter may be perceived as potential informal workers.

At this point, let us recall that by 2013, the urban economically active population is
3’182’187 people. We estimate that in the baseline scenario, the part of permanent
informal workers was 54.17% (1’723’791 people); this value is quite close to the
part of workers that have a level of education up to non-college technical studies
while not being affiliated with the pension funds, which is 55.36 % in the
Household Survey. We find that 2.97 percentage points of this part (94’511) can be
attributed to the 14th salary. Similarly, we find that 1.25 points (39’777) and 1.07
points (34’049) can be attributed to wage increases and contributions to “Fondo
Solidario”, respectively. We find that the 14th salary is the policy that has
contributed the most to the increase of permanent informal workers, followed by
wage increases and contributions to “Fondo Solidario”. We make the case that this
is due to the fact that, even if it is a contingent policy, the 14th salary represents the
highest monetary cost for the firm among all other considered policies. In the
absence of all three considered policies, the part of permanent informal workers
would be 48.88% (1’555’453), which is still high. Thus we argue that even if the
increasing costs for the formal firms have increased the attractiveness of the
informal sector, the size of the latter is the result of structural characteristics of the
Bolivian economy that transcend of labor policies only, and thus lie outside the
scope of this study.

We estimate that the part of potential informal workers in the baseline scenario
was 27.27% (867’782). We find that the 14th salary, minimum wage increases and
contributions to “Fondo Solidario” are the cause of 1.44 points (45’823), 0.77
points (24’503) and 0.66 points (21’002) of this part, respectively. We confirm that
among the three considered policies, the 14th salary contributes the most to the
expansion of the informal sector. To clarify this statement, it is worth noticing that
these potential informal workers find themselves, indeed, in the informal sector.
The probability for a medium productivity worker to be employed in the informal
sector is estimated around 94.5% in all the scenarios; the probability for this type
of workers to occupy a formal position is only 4.7%. These results reinforce the
argument that informality is a more complex structural problem in the Bolivian

17
economy associated to the to both the demand and supply sides of the labor
market. On the one hand, this figures show that formal positions are not available
even for workers with medium productivity who are forced to insert themselves in
the informal sector in the quest for a monetary income. On the other hand, the high
rate of potential informal workers that are actually informal may be the result of a
willingly taken choice motivated by individual preferences, culture and tradition.
Also, as stressed in Heckman & Pagés (2004) and Frolich et al. (2014), this
behavior may be induced by a perception of low expected gains of being a formal
worker and having access to short-term and long-term social security services,
due, for example, to lack of trust in these institutions.

The increase of informality due to the considered labor policies is reflected in the
reduction of the part of the workforce that would only consider taking up formal
positions as a source of income. We estimate that this part is 18.56% (590’614) in
the base scenario, which is close to the part of the workforce that have a level of
education higher than non-college technical studies while being affiliated with the
pension funds according to the Household Survey (16.25%). In the absence of all
the three labor policies we find that this part would rise up to 26.72% (850’280).
The difference of 8.16 percentage points can be divided in 1.73 points (67’215),
2.02 points (64’280) and 4.41 points (140’334) caused, respectively, by
contributions to “Fondo Solidario”, wage increases and the 14th salary. This
constitutes further evidence for the fact that the 14th salary has contributed the
most to the reduction of attractiveness of the formal sector compared to the
informal sector, followed by wage increases and contributions to “Fondo
Solidario”. We show that these policies have pushed highly productive workers
back to the category of workers of medium productivity, who, as stated before, end
up in the informal sector due to the lack of opportunities in the formal sector
and/or individual preference for informality.

It is interesting to notice that, besides increasing informality, the labor policies that
we consider have reduced unemployment rates. We reason that both effects are
intertwined in the sense that the abandonment of aspirations for formal positions
for some medium and highly qualified workers may depict easier conditions for
the unemployment rates to fall. We obtain an unemployment rate of 3.26%
(103’739) in the baseline scenario, which reproduces the official rate published by
the Bolivian government. In absence of all the considered labor policies, we
estimate that unemployment rate would have risen to 4.35% (138’425). In line
with data presented in introductory paragraphs, 81% of the unemployed
workforce in the base scenario is formed by high productivity workers, thus we
make the case that these figures show that the labor policies have augmented
opportunity costs of waiting for a chance of occupying a formal position while
being unemployed.

At this point, it is clear that the considered labor policies have contributed to the
increase of informality. However, another noticeable result of our study is that the
policies that imply an increase in the formal workers’ salary, namely the 14th salary
and minimum wage increases have also reconfigured wage schedules in a way that
results in a reduction of wage inequality within the formal sector. However, since
informal workers do not benefit directly of these wage increases in our model, a

18
corollary of this result is that the ratio informal wage/mean formal wage decreases
due to these policies. We make the case that this result shows that if informal
workers are completely deprived of the benefits of these policies, this ratio is
always bound to raise and threaten to exacerbate wage inequality between the two
sectors. In turn, contributions to “Fondo Solidario” also contribute to the reduction
of wage inequality within the formal sector, but they reduce the ratio informal
wage/mean formal wage.

In order to clarify these statements, let us focus first on the effects of 14th Salary
and wage increase. In a standardized continuum of wages, i.e. in the range 0-1, we
estimate that the mean wage in the formal sector in the baseline scenario is 0.5820
with a standard deviation of 0.0687. Without the 14th salary and minimum wage
increases, the mean wage would drop to 0.5548 while the standard deviation
would rise up to 0.0758. The increase of 0.0071 points in the standard deviation
can be split into 0.0050 and 0.0021, attributed to the first and latter policies,
respectively. This means that the 14th salary contributes the most to the reduction
of wage inequality within the formal sector. We argue that this may be due to the
fact these policies foster the concentration of a reduced number of highly
productive workers in the formal sector and facilitates the concentration of high
wages in this sector. In contrasting fashion, we find that the ratio informal
wage/mean formal wage would rise up from 0.62 in the base scenario to 0.63
without wage increases, and further up to 0.65 without the 14th salary. We stress
that this result may not be perceived as a measure of inequality between sectors,
because it relies on the relatively strong assumption of null effects (direct or
indirect) of labor policies over informal wages. Rather, we make the case that this
result warns about the effects of these policies in the extreme case in which this
assumption holds.

Let us now focus on the effect of contributions to “Fondo Solidario”. The fact that
this labor cost for the firm does not generate direct benefits for formal workers
implies that the mean formal wage reduces due to this policy (0.0078 points). This
may be due to the fact that this policy only implies costs for the formal sector and
not a redistribution of wealth between formal employers and formal workers, as is
the case of the other two policies. In other words, gains generated in the formal
sector are extracted with the potential to create benefits for agents that lie outside
this sector. Accordingly, this policy causes a reduction of the ratio informal
wage/mean formal wage (0.01 points), meaning that it has the potential to
contribute to close the gap between wages in the two sectors. Finally it is
important to notice, that similarly to the other two policies, contribution to “Fondo
Solidario” fosters the reduction of wage inequalities within the formal sector,
diminishing their standard deviation by 0.0041 points. This may be explained by
the fact that this policy implies further labor costs for the formal firms, thus
promoting the concentration of highly productive workers in the formal sector.

5. Concluding Remarks

In this paper we propose a practical framework for assessing the effects of three of
the most recent labor policies in Bolivia, namely the 14th salary, minimum wage
increases and contributions to “Fondo Solidario”, on the configuration of the urban

19
labor market. Our analysis focuses on the extent to which these policies affect the
sorting of workers between the formal and informal sectors, and how they modify
wages in the formal sector. The theoretical foundations of this framework are built
upon Albrecht et al. (2008), Diamond (1982) and Mortensen & Pissarides (1994),
and they consist of a job search and matching partial equilibrium model. One the
main virtues of this model is that it allows for the existence of a large informal
sector in the economy that is not treated as a parallel disadvantaged residual of the
formal sector; rather, the formal and the informal sectors are deeply intertwined
and workers may transit between them following optimizing behavior. Our main
results can be divided in three groups: the effects of the policies on the sorting of
workers between the formal and the informal sectors, on unemployment and on
formal wages schedule.

We are able to identify two types of workers in the informal sector: permanent and
potential. The first possess low levels of productivity that do not allow them to
aspire to occupy formal jobs; the latter possess medium levels of productivity and
may consider taking up informal jobs if they perceive that they are better off
choosing this option over waiting for the arrival of chances to occupy a formal
position. Both type of informal workers increase due to considered labor policies.
We estimate that by 2013, 1.7 million people (54% of the workforce) were
permanent informal workers; around 95’000 of these people are in this situation
due to the 14th salary, 40’000 due to the minimum wage between 2012 and 2013
and 35’000 due to contributions to “Fondo Solidario”. We also estimate that 0.87
million people (27% of the workforce) were potential informal workers of which
46’000 can be attributed to the 14th salary, 25’000 to minimum wage increases and
21’000 to contributions to “Fondo Solidario”.

We find that while increasing informality, these policies reduce unemployment

rates. In the absence of these labor policies, unemployed people could raise from
104’000 (3.26%) to 138’000 (4.35%).

The results that we present on the effects of these policies on wage schedules need
a bit more attention. We find that all three policies reduce wage inequality within
the formal sector; we argue that this is due to the concentration of highly
productive workers in the formal sector in presence of these policies. Interesting
differences between policies arise when assessing the effects on the ratio informal
wage / mean formal wage. The 14th salary and minimum wage increases promote a
reduction this ratio (0.03 percentage points); we argue that the reason for this
result is that these policies imply costs for employers but they also depict direct
increases of monetary compensations only for formal workers while leaving
informal wages unchanged. In contrasting fashion, contributions to “Fondo
Solidario” reduce the mean formal wage and thus increase the ratio informal wage
/ mean formal wage. We argue that this is due to fact that this policy implies a cost
for employers but do not imply direct monetary benefits for formal workers. These
results show that if the 14th salary and minimum wage increases do reach informal
workers, neither directly nor indirectly, they may increase the wage gap between
in the informal sector. In contrast, contributions to “Fondo Solidario” have the
potential to reduce this gap by promoting the redistribution of gains from the
formal sector to the informal sector.

20
We argue that the results that we present here are coherent and quite intuitive,
thus we believe that our main contribution through this study is twofold. First, we
prove that it is possible to capture the essential traits of the labor market in Bolivia
building on very sound theoretical and technical analytical instruments grounded
on search and matching models. The effects of labor policies can be assessed in a
very objective and logical manner within this framework, thus providing useful
tools for effective policymaking. Second, we present novel quantitative evidence
about the relation between labor policies and the configuration of the job market
in Bolivia, providing further inputs for a deep discussion about this socially crucial
matter.

We are well aware of the fact that such a socially sensible subject as the
configuration of job market cannot be assessed only on the grounds of economic
modeling techniques such as the ones that we present here. However, we defend
the idea that these analyses are of paramount importance towards more detailed
and precise discussions on this subject, as they provide logical and robust
frameworks that are exempt of personal judgments and political perspectives. We
intend to keep on undertaking studies in this sense and perfecting our analytical
tools, as we are convinced that they offer robust insights towards effective
policymaking for better jobs.

6. References

1. Albretch J., Navarro L. , Vroman S. (2009). The effects of labor market

policies in an economy with informal sector. The Economic Journal.
119(539): 1105-1129
2. Andersen, L., & Muriel, B. (2007). Informality and productivity in Bolivia: A
gender differentiated analysis. Development Research Working Paper Series
07/2007, Institute for Advanced Development Studies.
3. Bosch M. (2006). Job creation and job destruction in the presence of
informal labour markets. CEP Discussion Paper 761, London School of
Economics.
4. Bosch, M & Esteban-Pretel J. (2009). The informal labor market in Latin
America. Working Paper, University of Tokio.
5. Cosar, A. , Guner, N. & Tybout, J. (2013). Firm Dynamics, Job Turnover and
Wage Distributions in an Open economy. IZA Discussion Paper 7718.
6. Cox, D. R., & Oakes, D. (1984). Analysis of survival data Vol. 21. CRC Press.
7. Diamond, P. (1982). Wage determination and efficiency in search
equilibrium. Review of Economic Studies, 49, 217-27.
8. Frolich, M., Kaplan, D., Pagés, C., Rigolini, J. & Robalino, D. (2014). Social
insurance, informality and labor markets: How to protect workers while
creating good jobs. Oxford University Press, UK.
9. Gong, X., van Soest, A. & Villagómez, E. (2004). Mobility in the urban labor
market: A panel data analysis for Mexico. Economic Development and
Cultural Change, 53(1), 1-36.
10. Heckman, J. & Pagés, C. (2000). The cost of job security regulation: Evidence
from Latin American labor markets. NBER Working Papers 7773.

21
11. Heckman, J. & Pagés, C. (2004). Introduction to "Law and Employment:
Lessons from Latin American and the Caribbean, NBER Chapters, in Law
and Employment: Lessons from Latin America and the Caribbean, pp 1-108
National Bureau of Economic Research.
12. International Labor Organization (2013). Panorama Laboral 2013, América
Latina y el Caribe. OIT, Lima – Perou.
13. International Labor Organization (2014). Trends in informal employment in
Colombia: 2009-2013. Notes on Formality, Program for the Promotion of
Formalization in Latin America and the Caribbean.
14. Jiménez, E. (2003). Género versus Etnicidad: Un estudio de caso sobre
migrantes rurales y mercados de trabajo en Bolivia. Cuaderno de Trabajo 5,
Red Mujeres y Economía Bolivia.
15. Leal-Ordóñez, 2014. The informal sector in contemporary models of the
aggregate economy. Working Papers 2014-24, Banco de México.
16. Maloney, W. (2004). Informality revisited. World Development, 32(7), 1159-
1178.
17. Masatlioglu, W. & Rigolini, J. (2006). Informality traps XXX.
18. Mercado, A., Andersen, L. & Muriel, B. (2003). Discriminación étnica en el
sistema educativo y el mercado de trabajo en Bolivia. Revista
Latinoamericana de Desarrollo Económico 1, 63-98.
19. Mondragón, C. & Peña, X. (2008). Business ownership and selfemployment in
developing economies: The Colombian case. CEDE Working Paper 2008-03,
Universidad de los Andes.
20. Mortensen, D. & Pissarides, C. (1994). Job creation and job destruction in
the theory of unemployment. Review of Economic Studies. 61(3), 397-415.
21. Muriel, B. & Jemio LC. (2010). Mercado laboral y reformas en Bolivia in
Desafíos laborales en América Latina después de dos décadas de reformas
estructurales: Bolivia, Paraguay y Perú 1997-2008 (Eds. Rodríguez & Berry),
Departamento de Economía - Pontificia Universidad Católica del Perú.
22. Petrongolo, B & Pissarides, C. (2000). Looking into the black box: A survey
of the matching function. CEPR Discussion Papers 2409, London School of
Economics.
23. Robayo, M. (2014). Tax-transfer schemes, informality and search frictions in
a small open economy. Mimeo.
24. Román, S. (2011). Costos laborales, economía informal y reformas a la
legilación laboral en Bolivia. Development Research Working Paper Series
08/2011, Institute for Advanced Development Studies.
25. Saint-Paul, G. (1996). Dual labor markets. A macroeconomic perspective. MIT
Press.
26. Shapiro A. & Sarzosa M. (2013). Unemployment protection for informal
workers in Latin America and the Caribbean. IDB Working Paper Series No.
WP-385.
27. Shimer R. (2005). The cyclical behavior of equilibrium unemployment,
vacancies and wages: Evidence and theory. American Economic Review.
95(1), 25-49.
28. World Bank (2009). Increasing formality and productivity in Bolivia firms. A
World Bank Country Study 50358, The World Bank.
29. World Bank (2015). Doing business report 2015: Going beyond efficiency,
12th Edition, The World Bank.

22
7. Appendix

Solving for reservation productivity and productivity cuts

The reservation productivity schedule is endogenously determined in the

equilibrium, noting that if a worker performs at this level of productivity, the job is
on the verge of being destroyed. In this situation, there is no aggregated gain for
keeping the job:
𝑁1 (𝑅(𝑦̅), 𝑦̅) − 𝑈(𝑦̅) + 𝐽(𝑅(𝑦̅), 𝑦̅) + 𝑠 = 0 (10)

Solving (8), this means that:

𝑦̅
(𝑟 + 𝜆)𝑦̅((1 + 𝜏)𝑟𝑈(𝑦̅) − 𝑟𝑠) − 𝜆 (∫𝑅(𝑦̅)(1 − 𝑥)𝑑𝑥 − (1 − 𝑦̅)𝑦̅)
𝑅(𝑦̅) = (11)
𝑟𝑦̅ + 𝜆

Once the reservation productivity schedule is identifies, it is possible to identify

the level of productivity 𝑦 ∗ that separates informal workers from potential formal
workers. This level is determined in equilibrium and it is given by:

𝑦∗
𝑏(𝑟 + 𝛿) + 𝛼𝑦0 𝜆 𝜆
𝑦 ∗ = (1 + 𝜏) + ∗𝑠− ∗ ∫ (1 − 𝑥)𝑑𝑥 (12)
𝑟+𝛼+𝛿 𝑦 𝑦 (𝑟 + 𝜆)
𝑅(𝑦 ∗ )

Similarly, the level of productivity 𝑦 ∗∗ separting the formal workers from potential
informal workers is determined in equilibrium and it is given by:

𝑦∗
(1 + 𝜏)(𝑟𝑦 ∗∗ + 𝜆) 𝜆 𝜆
𝑦 ∗∗ = 𝑦0 + (1 + 𝜏)𝑦0 + 𝑠 − − ∫ (1 − 𝑥)𝑑𝑥 (13)
𝛽𝑚(𝜃)𝑦 ∗∗ 𝑦 ∗∗ 𝑦 ∗∗ (𝑟 + 𝜆)
𝑅(𝑦 ∗ )

Finally, the steady state conditions in which flows of workers moving between
states are constant, define the levels of unemployment for each type of worker as
well as the level of formal and informal employment. Let us note 𝑢(𝑦̅) the
probability for a worker of maximum level of productivity 𝑦̅ to be unemployed,
𝑛0 (𝑦̅) the probability of being employed in the informal sector and 𝑛1 (𝑦̅) the
probability of being employed in the formal sector. Thus 𝑛0 (𝑦̅) + 𝑛1 (𝑦̅) + 𝑢(𝑦̅) =
1 ∀𝑦̅.

For ‘low productivity workers’, the flow between unemployment and the informal
sector is constant (because they are never in the formal sector); for ‘high
productivity workers’ the flow between unemployment and the formal sector is
constant (because they are never in the informal sector, and for ‘medium
productivity workers’ the flow between unemployment and the formal sector as
well as the flow between unemployment and the informal sector are both constant.

These steady-state conditions define the following set of equations:

23
 𝛿
𝑢(𝑦) =
𝛿+𝛼
𝑖𝑓 𝑦 ≤ 𝑦 ∗ 𝛼 (14)
𝑛0 (𝑦) =
𝛿+𝛼
{ 𝑛1 (𝑦) =0

𝛿𝜆𝑅(𝑦)
𝑢(𝑦) =
𝜆(𝛿 + 𝛼)𝑅(𝑦) + 𝛿𝑚(𝜃)𝑦
𝛼𝜆𝑅(𝑦)
𝑖𝑓 𝑦 ∗ < 𝑦 < 𝑦 ∗∗ 𝑛0 (𝑦) = (15)
𝜆(𝛿 + 𝛼)𝑅(𝑦) + 𝛿𝑚(𝜃)𝑦
𝛿𝑚(𝜃)𝑦
𝑛1 (𝑦) =
{ 𝜆(𝛿 + 𝛼)𝑅(𝑦) + 𝛿𝑚(𝜃)𝑦

𝜆𝑅(𝑦)
𝑢(𝑦) =
𝜆𝑅(𝑦) + 𝑚(𝜃)𝑦
𝑖𝑓 𝑦 ≥ 𝑦 ∗∗ 𝑛0 (𝑦) = 0 (16)
𝑚(𝜃)𝑦
𝑛1 (𝑦) =
{ 𝜆𝑅(𝑦) + 𝑚(𝜃)𝑦

24